Robust Nonlinear PLS Based on Neural Networks and Application to Composition Estimator for High-Purity Distillation Columns

June Liu; Kwanggi Min; Chonghun Han; Kun Soo Chang

Korean Journal of Chemical Engineering, Vol.17, No.2, 184-192, March, 2000

June Liu, Kwanggi Min¹, Chonghun Han^{1, †}, and Kun Soo Chang¹

Mechanical Engineering Team, Samsung Engineering Co., Ltd, Korea
¹Department of Chemical Engineering, Automation Research Center, Pohang University of Science and Technology, Korea

E-mail:

The accurate and reliable on-line estimation of product quality is an essential task for successful process operation and control. This paper proposes a new estimation method that extends the conventional linear PLS (Partial Least Squares) regression method to a nonlinear framework in a more robust manner. To handle the nonlinearities, nonlinear PLS based on linear PLS and neural network has been employed. To improve the robustness of the nonlinear PLS, the autoassociative neural network has been integrated with nonlinear PLS. The integration allows us to handle the nonlinear correlation as well as nonlinear functional relationship with fewer components in a more robust manner. The application results have shown that the proposed Robust Nonlinear PLS (RNPLS) performs better than previous linear and nonlinear regression methods such as PLS, NNPLS, even for the nonlinearities due to operating condition changes, limited observations, and measurement noise.

Keywords:PLS;Neural Networks;Nonlinear Correlation;Composition Estimator;Distillation Column

[References]

Baratti R, "Composition Estimators for Distillation Column : Theory and Experiments," DYCORD '95, Copenhagen, Denmark, 475 (1995)
Baratti R, Bertucco A, Darold A, Morbidelli M, Chem. Eng. Sci., 50(10), 1541 (1995)
Baratti R, Hydrocarb. Process., 74(6), 35 (1995)
Dong D, Comput. Chem. Eng., 20, 65 (1996)
Holcomb TR, Comput. Chem. Eng., 16, 393 (1992)
Hornik K, Neural Networks, 3, 551 (1990)
Hoskuldsson A, J. Chemometrics, 2, 211 (1988)
Joseph B, AIChE J., 24, 485 (1978)
Kiparissides C, "Real-time Optimization and Model-based Control of Polymer Reactors," Methods of Model Based Process Control, Berber, R., eds., Kluwer Acdemic Publisher, Netherlands, 495 (1995)
Kramer MA, Comput. Chem. Eng., 16, 313 (1992)
Kramer MA, AIChE J., 37, 223 (1991)
Kresta JV, Comput. Chem. Eng., 18, 597 (1994)
Ljung L, "System Identification-Theory for User," Prentice Hall, Englewood Cliffs, NJ (1987)
Martens H, "Mutivariate Calibration," John Wiley & Sons, New York (1989)
Mejdell T, Ind. Eng. Chem. Res., 30, 2555 (1991)
Mejdell T, Ind. Eng. Chem. Res., 33, 2543 (1991)
Mejdell T, AIChE J., 39, 1641 (1993)
Orfandis SJ, Neural Computation, 2, 116 (1990)
Palus M, Physica D, 55, 221 (1992)
Piovoso MJ, "Sensor Data Analysis using Artificial Neural Networks," Int. Conf. Chem. Process Control, CPC IV, Texas (1991)
Qin SJ, Comput. Chem. Eng., 16, 379 (1992)
Rumelhart DE, "Parallel Distributed Processing, Explorations in the Microstructure of Cognition, Vol. 1: Foundations," MIT Press, Cambridge (1986)
Scales LE, "Introduction to Non-Linear Optimization," Springer-Verlag, New York (1985)
Skogestad S, Ind. Eng. Chem. Res., 27, 1848 (1988)
Soderstrom T, "System Identification," Englewood Cliffs, New Jersey (1989)
Su H, Ind. Eng. Chem. Res., 32, 1927 (1993)
Wold S, Technometrics, 20, 397 (1978)
Xu L, Neural Networks, 5, 441 (1992)

[Referenced By]

[Related Articles]

[1] Baratti R, "Composition Estimators for Distillation Column : Theory and Experiments," DYCORD '95, Copenhagen, Denmark, 475 (1995)

[2] Baratti R, Bertucco A, Darold A, Morbidelli M, Chem. Eng. Sci., 50(10), 1541 (1995)

[3] Baratti R, Hydrocarb. Process., 74(6), 35 (1995)

[4] Dong D, Comput. Chem. Eng., 20, 65 (1996)

[5] Holcomb TR, Comput. Chem. Eng., 16, 393 (1992)

[6] Hornik K, Neural Networks, 3, 551 (1990)

[7] Hoskuldsson A, J. Chemometrics, 2, 211 (1988)

[8] Joseph B, AIChE J., 24, 485 (1978)

[9] Kiparissides C, "Real-time Optimization and Model-based Control of Polymer Reactors," Methods of Model Based Process Control, Berber, R., eds., Kluwer Acdemic Publisher, Netherlands, 495 (1995)

[10] Kramer MA, Comput. Chem. Eng., 16, 313 (1992)

[11] Kramer MA, AIChE J., 37, 223 (1991)

[12] Kresta JV, Comput. Chem. Eng., 18, 597 (1994)

[13] Ljung L, "System Identification-Theory for User," Prentice Hall, Englewood Cliffs, NJ (1987)

[14] Martens H, "Mutivariate Calibration," John Wiley & Sons, New York (1989)

[15] Mejdell T, Ind. Eng. Chem. Res., 30, 2555 (1991)

[16] Mejdell T, Ind. Eng. Chem. Res., 33, 2543 (1991)

[17] Mejdell T, AIChE J., 39, 1641 (1993)

[18] Orfandis SJ, Neural Computation, 2, 116 (1990)

[19] Palus M, Physica D, 55, 221 (1992)

[20] Piovoso MJ, "Sensor Data Analysis using Artificial Neural Networks," Int. Conf. Chem. Process Control, CPC IV, Texas (1991)

[21] Qin SJ, Comput. Chem. Eng., 16, 379 (1992)

[22] Rumelhart DE, "Parallel Distributed Processing, Explorations in the Microstructure of Cognition, Vol. 1: Foundations," MIT Press, Cambridge (1986)

[23] Scales LE, "Introduction to Non-Linear Optimization," Springer-Verlag, New York (1985)

[24] Skogestad S, Ind. Eng. Chem. Res., 27, 1848 (1988)

[25] Soderstrom T, "System Identification," Englewood Cliffs, New Jersey (1989)

[26] Su H, Ind. Eng. Chem. Res., 32, 1927 (1993)

[27] Wold S, Technometrics, 20, 397 (1978)

[28] Xu L, Neural Networks, 5, 441 (1992)